About a year ago, I shared the first post in what will hopefully be a long series about starting and growing a company. The timeline went something like this:
- April 2011 – started the company.
- April 2012 – year one finished, but I didn’t blog about it right away. I wanted to get it into the rear view mirror for perspective before blogging.
- April 2013 – year two finished, so I blogged about year one.
- April 2014 – year three finished, so I’m blogging about year two now.
When Last We Met
As we finished up year 1, we had just brought Employee #1, Jes Schultz Borland, on board. Jeremiah, Kendra, and I were doing different variations of our health check process, and we had to get things to be more standardized.
We invested big, pouring five figures of cash into rebranding our web site, implementing SalesForce.com, and becoming an Amazon partner. We didn’t want to be a big, faceless corporation, but we wanted to start growing our product range and our presence online.
So how’d year two go?
The New Web Site: Amazing.
We hired PixelSpoke to design our brand – our logo, our web site, our font, etc. They started by asking us (and some of our clients) all kinds of interesting questions. They built a profile of how we saw ourselves, how our customers saw us, and how we wanted to be seen.
Three words: loving commando nurses.
I am not making this up.
Before PixelSpoke, BrentOzar.com was an off-the-shelf WordPress theme that felt like a personal blog, not a consulting company that supported three people full time. We didn’t want to masquerade as Global Megacorp Consulting – we wanted something that felt really hand-crafted and personal with as few straight lines as practical. They nailed it.
PixelSpoke introduced us to Eric Larsen, the artist who drew our portraits for the site, and continues to draw portraits for our slide decks and promotional materials.
We had some bumps along the way – for example, our first logo contained a red cross, the only logo protected by the Geneva Convention. Oops. We discovered this stuff while working with our (awesome) lawyers to protect our brand name, our logo, and our scripts.
The Legal Stuff: Covering Our Butts
We embarked on a mission to trademark and copyright the stuff that we were working really hard to build. At first I felt kind of guilty – after all, I love giving away stuff to the community – but we had to make sure it didn’t get abused.
For example, during year 2, I got an email from a software company giving me a polite heads-up that they were including sp_Blitz® in the next version of their product. They thought they were doing us a favor by getting it more publicity, but that immediately freaked me out. I had visions of “Powered by Brent Ozar’s sp_Blitz” in their marketing material, and I politely asked them to hold off.
To protect ourselves, we trademarked and copyrighted stuff, built an End User License Agreement, and forced users to agree to the EULA before downloading scripts like sp_Blitz® and sp_BlitzIndex®.
A little part of me died inside when we did that, but it was the right thing to do. Our babies were growing up. (Even now, in spring 2014, our admin is still adding ® marks to our trademarked terms scattered around the web and in our documents.)
We also stopped hosting community bloggers on our internet host. For years, I’d personally picked up the tab to host dozens of community blogs including Kevin Kline, Jorge Segarra, Erin Stellato, Ryan Adams, Tim Ford, and Michael Swart. I’d originally started doing that because I was maniacally dedicated to fostering new blogger voices in the SQL Server community, and I didn’t want anything to stand in their way – even a $10/mo hosting bill. Many of these bloggers had grown up to MVP status, publishing books, and even going to work for competing consulting companies. With us moving our hosting into Amazon EC2, it just seemed like the right time to encourage those bloggers to leave the nest and get their own hosting. It broke my heart a little, but those folks were grown up too, and I’m so proud to see them continuing to blog today.
The Web Site Made Sales Way Easier
The single toughest thing about year one of our company was the sales process. Oh, wow, did it suck. We had no idea how to communicate with new clients who kept saying the same thing over and over: “We want to work with Brent.”
We tried communicating the strengths of the other team members. We tried keeping me out of the sales calls. We tried raising my rates above the rest of the staff. We tried pushing my availability farther out into the future. We tried saying I wasn’t accepting new clients. Nothing seemed to work.
At some point in year 2, the problem simply disappeared. I can’t pinpoint exactly when, but I think it was due to a few different factors:
- The new web site felt much more like a company, not a lone guy blogging about stuff.
- Each post had clear bylines with pictures of the authors
- Jeremiah and Kendra had built up a solid history of blog posts, and they were bringing in search results for things that I’d never blogged about
- The contact page made it clear that multiple people were involved in the company
- We made a new pitch deck for sales calls that emphasized our process
The end result: new clients stopped demanding that I work on their project, and started happily working with any of us.
This change came with a wonderful byproduct – each of us had distinct roles in the company that fit our personalities really well. I do sales and marketing, Kendra manages people and processes, and Jeremiah is research & development.
Became an Amazon Web Services Partner
In mid-2012, we believed that Amazon Web Services held a lot of promise for us. I believed (and still do) that businesses would rather pay by the hour for flexible IT infrastructure than make huge capital outlays. After all, we’re a small business and we don’t own a single server either.
To join the Amazon Web Services Partner Network, we had to shell out some cash, pass a few tests, and two of us had to attend Amazon’s architecture training classes. Jeremiah and I went, and I enjoyed it a lot. We set up our Amazon Partner Network profile page, set up pages on our own site advertising our AWS consulting, and blogged about some of the work we’d done. We hosted our web site in a load-balanced, somewhat-geographically-redundant EC2 setup.
This experiment cost us in the neighborhood of $30k-$50k.
And nearly no business came in. (Certainly not $30k-$50k worth.)
We learned a funny lesson that seems obvious in retrospect; you don’t need a consultant to get SQL Server into Amazon Web Services, just as you don’t need a consultant to get it installed in on-premises bare metal. You need a consultant after you’ve already gone live and the performance or reliability isn’t what you expect.
We’re not a VMware, NetApp, EMC, or Dell partner either, but our blog posts about running SQL Server on those platforms mean that people call us for slow SQL Server performance. We realized that Amazon Web Services isn’t the central part of our offering, but our expertise on the subject is just one of the reasons people call us. We don’t need to be an official partner in order to get that business.
People don’t find us by searching vendor partnership directories.
People find us by searching Google.
Later on in year 3, this realization helped us crystallize our product offerings down to just one product: our SQL Critical Care®.
Other Year Two Bets
The $30k-$50k AWS bet didn’t feel stressful or wasted – it seemed like the right bet at the time, and I don’t think any of us ever second-guessed it. I’m telling the story here to illustrate what it’s like starting a small business: some (many? most?) of your bets will not pay off, and that’s okay.
Some of the bets we made in year two included:
- Investing in SalesForce.com to track downloads and client leads (total failure because we didn’t have enough manpower to follow up on the leads, and we didn’t want to pester them)
- Hiring Lori Edwards to make improvements to sp_Blitz® (success because we didn’t have enough time to make a lot of necessary fixes ourselves)
- Hiring a contractor to build an sp_Blitz® app (success, because people continue to use it to this day, but kind of a pain to keep up to date)
- Hosting our second company retreat, this time in Cabo, Mexico (success, because chilaquiles and donkeys)
Refocused My Speaking Efforts
I was honored to be invited to present Building the Fastest SQL Servers at both Microsoft TechEd North America and in Europe. You can watch the screencasts of the Orlando session and the Amsterdam one, and they were both a ton of fun.
I was flabbergasted when the live attendees voted me into the top 5 sessions at both conferences (out of hundreds of sessions), up alongside incredible speakers like Mark Russinovich and Mark Minasi. That was a real gut-check moment. For years, I’ve been accustomed to having really high ratings in the SQL Server community, but to be rated one of the best sessions across Windows, development, and SharePoint tracks? Wow, okay, maybe I’d gotten good at this stuff.
Despite the rush of really good feedback, I decided not to speak at TechEd again. The way the conference was set up, attendees all thought I was a Microsoft employee, and that wasn’t good for my branding. Presenters actually lose money at these types of events, and the only way to break even is to gain consulting or training business. That doesn’t work when attendees think you’re a blue badge. (If I’m going to give away training and lose money, I’m going to do it for a community-driven conference, not a vendor event.)
I came to the same conclusion about company internal training events, too. I was really excited to be invited to speak at a Fortune 100 company’s annual learning conference, and I thought it’d get my foot in the door for additional revenue. Nope – even though the employees had a great time at the talks, it didn’t turn into consulting work because big companies already have preferred vendor relationships with big consulting companies.
It’s really cool to be invited to these big events – but if they don’t produce revenue for the company, is it really good for me (and the company)? I love helping people, but in year 2, I learned that I need to focus on a couple of simple questions:
- Are we going to get revenue (consulting or training) enough to offset the trip?
- If not, how can we maximize the number of people we help at the lowest cost to everyone involved?
- How can we control the quality of the event?
We wanted to make as many successful bets as possible, which brought us to a big milestone:
Ran Our First Training Class
When I attended conferences as an employee, I thought to myself, “I really need a guided list of sessions that I should attend. I only want to attend the best stuff for people like me.” I didn’t want to sift through a bunch of really crappy unrelated sessions. I didn’t want to wonder who the good speakers were and who was going to just read off their slides. I wanted a clear progression of sessions that built upon each other and taught me real-world useful stuff. I wanted a summary module that drove it all home, connected the dots, and told me what to do when I got back to the office.
Except I didn’t know I wanted all that until Jeremiah, Kendra, and I sat down and talked about what sucked about conferences, and how we could build something different.
We started with a 2-day developer training class in Atlanta in spring 2013. We were already used to doing 1-day pre-conference sessions at major conferences, and we figured a 2-day class was the next logical step up.
At the time, it felt like a huuuuge risk. We were committed to running the class even if just a couple of attendees showed up. We had to shell out tens of thousands of dollars for travel, hotel, banquet room fees, and advertising. We blocked out weeks on our calendar to build, test, and rehearse the class material.
Sure, our training-class-at-sea SQLcruise idea had panned out, but similar efforts from other community members didn’t catch on. What if the cruise thing was just luck? And sure, my online training sales were going fairly well – I was running live virtualization and tuning classes at $200-$300/pop with pretty good turnout, but many of those sales were international attendees. Would people actually fly to a class in Atlanta?
Thankfully, it worked out – around 30 people went for it, and the event finished out our year 2 with a success. At least one of our bets paid off.
What I Expected for Year Three
Looking at my calendar and my to-dos as of April 2013, one thing was looming large: training.
Our in-person training at conferences was getting rave reviews. Companies were paying us to bring the training onsite, and companies were willing to fly their employees to a training event in Atlanta. We needed to continue to ramp this up in year three.
Our remote video training was successful too, but was a total nightmare to operate. It involved a hacked-up combination of EventBrite, WebEx, and a WordPress plugin. We wanted to scale this up too, but it had to be higher quality for both the attendee and the organizer.
What I (Not We) Screwed Up in Year Two
In retrospect, there’s one thing I did a really, really crappy job on: accounting. Sure, we had an accounting firm managing the books, and they were all in order, but we weren’t paying much attention to the details. We just kept an eye on the bank balance and made sure it kept going gradually up.
At one point in a partner discussion, I sat bolt upright and realized I had spent the first part of my career closely reviewing budgeting and P&Ls for someone else’s hotels, and I had simply not even glanced at that stuff for our very own business. Hello. What a total lapse of judgment. I couldn’t even begin to tell you how much money we’d spent on a particular conference, what our overall profit rate was on training or consulting, or whether our first employee was profitable. I was a data guy, for crying out loud, and I didn’t even have my finger on the most basic statistics of our business.
I say I screwed this one up because this is something I know how to do, and I’ve done before throughout my career. I know better, and I should have done better. As a company, we made pretty good decisions on when to make additional bets and when to stop putting money down, but we could have been much more informed when we made those bets. That’s my fault, and it’s something I still struggle with today.
Making Decisions About Time Allocation
It’s tough as a founder because I’ve only got so many hours in the day.
Right now, for example, it’s 9:45AM on Saturday morning. I spent about an hour this morning going through emails and putting some infrastructure in place for our 2015 training classes, and then I spent an hour and a half writing this post.
Should I have been reviewing the P&L instead, looking at how our employee profitability is going, and projecting the date when we can hire employee #3? Should I have finished the expense reports for my recent SQLSaturday Lisbon and SQLintersection Orlando trip? Or should I just surf Reddit? I know I’ve only got about another 12 minutes before Erika wakes up, and from that point on, the rest of our Saturday will be spent together. (To be a good partner, I try not to work on weekends when she’s awake.)
I chose to put the time into this post because writing and sharing rejuvenates me. A long time ago, Tom LaRock told me something that really stuck with me: if you want to stay motivated, you’ve gotta spend your true spare time doing the things that energize you, and avoid doing things that drain you. It’s all a tricky balance of doing enough fun stuff to keep you motivated to do the hard work that pays off later.
In year two, time allocation was probably my toughest challenge. Erika and I made a deal that I’d travel no more than one week per month. My quality of life went up after that, I think, and it forced me to think more about video sales and passive income. As much as I love travel and speaking, I had to cut back in order to keep making progress on all of my Epic Life Quest goals and not just focus on a single area.